Research on optimization of main steam pressure of units based on real-time cost analysis
WANG Huijie
HAN Yu
GAO Haoran
XU Xiaogang
Abstract：Based on the historical operation data and business data of thermal power units,a real-time cost prediction model for thermal power plants is established by deep neural networks(DNN).Through the improved dung beetle optimization algorithm(I-DBO),the initial pressure optimization of the model is carried out within the feasible range of the main steam pressure.The optimized operation curve is compared with the original curve of the manufacturer,and the impact of the fluctuation of standard coal prices on the real-time cost is analyzed.The results show that after optimization,the real-time cost of the unit under each load has been significantly reduced.Power plants can participate in the real-time quotation of the power grid based on real-time costs and formulate personnel assessment standards.
Keywords：thermal power plantreal-time costoptimal initial pressuredeep neural networkdung beetle optimization algorithm
Publication Date：2025-07-25
Online Publishing Date：2025-09-15(First online date of this platform, not the publication date of the document)
Pages：6( 70-75 )